The long-term objective of this project is to develop the optimal multi- modality approach to local breast cancer staging. Currently, the methods used to determine the extent of cancer in the breast involve palpable findings, mammography, and pathologic analysis of biopsy. Often, multiple excisions are necessary to establish the extent of cancer within a breast. In addition, disease remaining in the breast often goes undetected.. This study will investigate the ability of the new generation of breast imaging modalities to determine the extent of breast cancer in the affected breast and distant sites (for PET). In the current proposal, women with newly diagnosed breast cancer will undergo a battery of imaging tests consisting of digital mammography, ultrasound, PET, and MRI prior to definitive surgery in an attempt to determine the extent of the index lesion, the presence of additional foci disease not detected by initial mammography and physical examination and the presence of nodal and distant metastasis. Specific aims of this proposal are: Aim 1: Compare the ability of digital mammography, ultrasound, MRI and PET in the depiction the extent of the index cancer within an affected breast. Aim 2: Compare the ability of digital mammography, ultrasound, MRI and PET in the identification of additional foci of cancer. Aim 3: Compare the accuracy of the various imaging examinations in predicting the status of the axillary lymph nodes. Aim 4: Determine whether various imaging examinations in predicting the status of the axillary lymph nodes. Aim 4: Determine whether diffuse or multi-focal cancers detected on MRI, ultrasound, digital mammography and PET can be predicted based mammographic, clinical, or pathologic findings. Aim 5: Based on correlation of various imaging techniques, clinical, and pathological findings generate a profile of patients at risk for mammographically and clinically occult diffuse or multi-focal disease. In order to investigate the predictive power of these mammographic and clinical features, we will correlate the imaging results from all modalities (and pathologic results of image guided biopsy) with the mammographic and clinical findings which lead to the detection of the initial cancer. It is hypothesized that this study will permit us to identify those cases where MRI, PET, digital mammography and ultrasound are most likely to detect diffuse or multi- focal disease, and thereby alter management. We plan to enroll 400 patients into the study and the data generated will determine the sensitivity, specificity and other appropriate parameters related to each modality employed.